论文部分内容阅读
快速准确无损测定种子生活力是种质资源安全保存研究中的一项重要内容。采用傅立叶变换近红外漫反射光谱技术,以保存不同年限的大豆品种资源的种子为样品,利用偏最小二乘法(PLS)建立其生活力的无损测定校正模型,并对不同光谱预处理方法和不同建模波段范围对模型的预测性能进行对比分析。结果表明:利用4000~10000/cm全波段建立的模型预测性能较高。经Savitzky-Golay二阶导数和标准化预处理后,生活力的PLS模型预测性能最好,校正集的决定系数为0.937,验证集决定系数为0.902,校正均方根误差(RMSEC)和验证均方根误差(RMSEV)分别为2.190和2.684。模型预测误差接近标准发芽方法测定误差,说明该方法准确性较高,能够满足种质资源快速、非破坏性生活力检测的要求,为今后快速无损测定种子生活力提供了方法。
Rapid, accurate and non-destructive determination of seed vigor is an important part of the research on the safe preservation of germplasm resources. Using Fourier Transform Near Infrared Diffuse Reflectance Spectroscopy (RS-NIRS), a non-destructive calibration model was established using partial least square method (PLS) to preserve the seeds of soybean cultivars with different years of age. The different spectral pretreatment methods and different The range of the modeled band is compared with the predicted performance of the model. The results show that the model established with 4000 ~ 10000 / cm full band prediction performance is high. After Savitzky-Golay second-order derivative and normalized pretreatment, the viability PLS model has the best prediction performance, the calibration coefficient of the calibration set is 0.937, the validation set determination coefficient is 0.902, the root mean square error of correction (RMSEC) and the mean square error The root-mean-square error (RMSEV) was 2.190 and 2.684, respectively. The prediction error of the model is close to the error of the standard germination method, which shows that the method has high accuracy and can meet the requirement of rapid and non-destructive testing of germplasm resources, and provides a method for fast and nondestructive determination of seed viability in the future.